Historical Value at Risk (VaR) with Python

[@oldtimehockey7123]

Thanks for sharing all of this information and code. The hill is steep, but worth it, I feel.

[@Saboman]

If i have the historical daily var and i want to comvert it to a 30 day period window, i just multiply by the sqrt(30)??? Thank you very much.

[@mediumdatascience8610]

Hey man! awesome video, cleared a lot of concepts. I have a doubt. If I wnated to calculate each day VaR or CVaR in a sliding/rolling window method with window size=7, and want to get each day's VaR or CVaR of each stocks or Portfolio, how to do that?

[@denisplaj6498]

I am not sure if we are allowed to simply multiply daily returns by T. They are not log returns, so wouldn't it be more correct to do something like this: (1+returns)**T - 1?

[@ali-claudebarry4076]

Hi mate, Great work and thanks for sharing, I’m struggling to apply your code based on an Excel file where I simply have portfolio monthly returns. I get “can’t multiply sequence by non int of type float” . I’m not sure how to approach this, I’m sure it’s because of the first steps but I’m quite new to Python so any help would be more than welcome. Cheers

[@finvengers6828]

Great Video .. if possible please make more videos on credit risk and FI derivatives valuation

[@gian_piano]

thank you so much for the video, I am going through your videos! I would like to become a quant that is my dream job

[@krishnakundu]

Hi…when i am giving the last code u did with the initial investment part..i got an error: cannot pack non-iterable Nonetype object and i am basically getting it in the code: pRet, pStd = PortfolioPerfomance(weights, meanReturns, covMatrix, Time)..can you pls help me with this?

[@mr.gk5]

Could you elaborate on the weights, why is it random? what if I want to put in my own weights?

[@QuantPy]

That would be fine 👍

[@AkashChauhan-rw1qh]

Thanks, You so much mate.......... It is very useful

[@patite3103]

Amazing video!

[@Im-Assmaa]

Thank you for the video. I have a question. How can I compute the quantiles for a specific p, using Rankit-cleveland method? It is used to estimate the value at risk using quantile regression and I am kind of stuck. please help

[@joker2117-f5v]

Actually I have a question, why do we need to add stock+'.AX'?

[@WorldWideSk8boarding]

this is really good

[@tblaze5406]

Hey mate, is the historical Var just the n th percentile of the profit loss distribution?

[@QuantPy]

Yes, otherwise thought of as the required capital to guarantee your portfolio or position is above 0 to that confidence interval.

[@Vince-rm7rr]

Hello, here the code is used with random weight. But what would be the code with lets say a 16.66% weight for each of your stock as you have 6 stocks ?

[@QuantPy]

just create a weights vector of your assigned weights and use that instead? w = np.array([1/len(stocks) for n in stocks])

[@nccamsc]

Being picky here: The Pythonic function name would be get_data, not getData, per PEP-8

[@TheChievovr]

very good!

[@alfilart]

Hi mate, First of all thanks for this helpful VaR series. I tried to run the code, but in the function historicalCVaR , I get an error in the line: belowVaR = returns

[@TheG0ldx]

That’s because you named a variable (the one where you store your series) the same as your function, historicalVaR in this case

[@aarondelarosa3146]

Excellent. You forgot to plot the chart.

[@user-wr4yl7tx3w]

may be breaking changes, I get the following error, RemoteDataError: No data fetched using 'YahooDailyReader'

[@mathysferriere3987]

from pandas_datareader import data as pdr import yfinance as yf yf.pdr_override()

[@mariaaguilera56]

I get: "RemoteDataError: No data fetched using 'YahooDailyReader'". I checked the symbols and they are correct.

[@mathysferriere3987]

from pandas_datareader import data as pdr import yfinance as yf yf.pdr_override()